Anzeige
Erschienen in:
2018 | OriginalPaper | Buchkapitel
Non-obtrusive Sleep Detection for Character Computing Profiling
verfasst von : Alia ElBolock, Rowan Amr, Slim Abdennadher
Erschienen in: Intelligent Human Systems Integration
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Abstract
The majority of existing Adaptive Systems rely on the user’s current state (affect) without taking the user’s general state (character) into consideration. In order to achieve truly seamless adaptive interactive systems, understanding the user’s character (i.e. Character Profile) is required. This paper presents a non-obtrusive sleep detector, MySleep, which is part of a multimodal lifelogging platform called MyLife. MyLife is designed for the main purpose of enabling building Character Profiles for users, which is a main artefact required in Character Computing. The aim of MySleep is to provide sleep records to be used in character profiling without requiring the user to use any external hardware and with minimal interaction. A study was conducted to test the accuracy of MySleep and compare it to other wearable sleep detectors. For the required purposes, the results provided by MySleep are accurate enough with requiring minimal interaction with the user.